Efficient rib plane visualization via automatic enhancement

Abstract

And the rib plane visualization based on the axial view is enhanced. Data representing a three-dimensional diagnostic image is received. Ribs are segmented, and rib centerlines are detected and marked. For each rib, a fitting plane corresponding to the rib centerline is calculated, respectively, and an out-of-plane rib portion is determined. Visualization of ribs is generated individually or as rib pairs. An enhanced visualization is generated for the out-of-plane rib portion, and the enhanced visualization may be performed via a fusion scheme or a curved surface scheme. The fusion scheme includes projecting an out-of-plane rib portion onto a visualization, such as fusing with an in-plane rib portion. The curved surface scheme includes fitting a polynomial surface such that all points in the set of points will be close to the surface. With this approach, individual ribs (pairs) can be easily examined in a single axial view.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging, and in particular to a method and apparatus for rib plane visualization via automatic enhancement. Background Art

[0002] Reading imaging scans, and more specifically trauma or emergency department scans, is a time-critical task that needs to be done with great care to avoid overlooking key findings. Often, in these settings, imaging is based on whole-body scans, which results in a large amount of imaging data being generated and which must be thoroughly reviewed. The ribs are particularly critical structures to assess (e.g., for trauma). These are repeated structures (typically 12 rib pairs and 24 vertebrae) that take a lot of time to review.

[0003] Computed tomography (CT) scanning has become the modality of choice for assessing the overall condition of a patient for applications such as trauma or emergency situations. Medical personnel such as radiologists often rely on medical image data in the form of CT scans for diagnostic purposes, such as the detection of rib fractures. During reading, each of the 24 ribs needs to be followed individually as it scrolls through the image slices.

[0004] Reviewing a chest or whole body three-dimensional (3D) CT scan slice by slice is often a time-consuming process, especially when the target anatomy spans multiple slices (eg, 24 individual ribs to be tracked). As a result, rib abnormalities may be overlooked.

[0005] The evaluation of ribs requires a considerable amount of reading time, since the ribs are usually tracked one by one through the volume of image data. Furthermore, pathologies such as fractures, especially buckles, can be very subtle and easily overlooked.

[0006] Several methods have been proposed to simplify the visualization and assessment of a patient's anatomy and critical structures, particularly those directed to visualization of the thoracic cavity. While offering some benefits, current methods have significant fundamental limitations.

[0007] One well-known visualization scheme is the "filet view" or "fishbone view". This view is based on segmenting the ribs (e.g., using a deep convolutional neural network), followed by a centerline extractor that subsequently labels the rib pairs in the field of view. Each rib is sampled along its trace, allowing visualization of each rib in a normalized and straightened manner (curved plane reformatting). This view allows a medical professional (such as a radiologist) to accurately inspect the rib centerlines in a normalized view, where all ribs are straightened and placed in a unique position on the inspection canvas (reformatted view).

[0008] However, this type of view has several disadvantages. One disadvantage of this type of view is that the nature of processing each rib independently results in discontinuities between ribs, which can cause imaging artifacts from adjacent ribs to appear in the rib shape.

[0009] Another visualization scheme is the visceral cavity view. In this view, a segmentation algorithm (e.g., using a model-based approach) is applied to segment the interior of the thorax based on the deformation of a cylindrical manifold. Once the segmentation is completed, the manifold can be unfolded and the maximum intensity projection (MIP) close to the surface can be calculated. This view allows the user to examine the thorax as a whole based on continuous visualization on the inspection canvas.

[0010] One of the drawbacks of this type of view is that the relative rib lengths are not maintained. In other words, the properties of the unfolded cylindrical manifold do not allow visualization of the correct rib lengths. For example, the first rib appears too long relative to the other ribs. In addition, the properties of MIP visualization may not allow detection of subtle fractures. For example, MIP visualization may make small rib fractures invisible and undetectable in the generated view. Another drawback is that the view adds significant unrealistic distortion (wavy ribs), which limits the clinical confidence of the generated visceral cavity views.

[0011] Therefore, there is a need for an effective rib plane visualization that overcomes the disadvantages associated with these views. Summary of the invention

[0012] The present invention aims to provide a technique for automatically enhancing views that overcomes the deficiencies in existing visualization schemes. The technique can be applied to multiple imaging systems, including CT, CT arms, single photon emission computed tomography CT (SPECT-CT), magnetic resonance CT (MR-CT), positron emission tomography CT (PET-CT), and magnetic resonance imaging (MRI) systems.

[0013] According to a first aspect of the present invention, a method for enhancing rib plane visualization is provided. The method comprises: receiving data representing a three-dimensional diagnostic image, the three-dimensional diagnostic image comprising one or more ribs of an object; segmenting the one or more ribs according to the received data representing the three-dimensional diagnostic image; detecting and marking rib centerlines according to the rib segmentation; calculating a fitting plane corresponding to the rib centerline for each rib respectively; determining an out-of-plane rib portion for each rib; generating a visualization of the ribs individually or as a rib pair; and generating an enhanced visualization of the out-of-plane rib portion.

[0014] In a second aspect of the present invention, an imaging device is provided. The imaging device includes: a memory configured to store computer executable instructions; and at least one processor configured to execute the computer executable instructions to enable the imaging device to: receive data representing a three-dimensional diagnostic image, the three-dimensional diagnostic image including one or more ribs of an object; segment the one or more ribs according to the received data representing the three-dimensional diagnostic image; detect and mark rib centerlines according to the rib segmentation; calculate a fitted plane corresponding to the rib centerline for each rib separately; determine an out-of-plane rib portion for each rib; generate visualization of the ribs individually or as a rib pair; and generate enhanced visualization of the out-of-plane rib portion.

[0015] In a third aspect of the present invention, a non-transitory computer-readable medium is provided on which instructions are stored for causing a processing circuit to perform a process. The process includes: receiving data representing a three-dimensional diagnostic image, the three-dimensional diagnostic image including one or more ribs of a subject; segmenting the one or more ribs according to the received data representing the three-dimensional diagnostic image; detecting and marking rib centerlines according to the rib segmentation; calculating a fitted plane corresponding to the rib centerline for each rib respectively; determining an out-of-plane rib portion for each rib; generating a visualization of the ribs individually or as a rib pair; and generating an enhanced visualization of the out-of-plane rib portion.

[0016] In a preferred embodiment, generating an enhanced visualization for the out-of-plane rib portions includes calculating a distance to the fitting plane for each rib centerline point, and wherein the rib portions outside the fitting plane are projected onto the visualization of the ribs, as fused with the in-plane rib portions. The out-of-plane rib portions may be highlighted to identify rib portions that are not part of the original visualization of the ribs. One aspect of the invention provides for switching between the original visualization of the ribs and the enhanced visualization of the ribs, and highlighting the radius of influence in the enhanced visualization of the ribs.

[0017] In a preferred embodiment, generating an enhanced visualization for the out-of-plane rib portion includes: establishing a point set including all centerline points of the rib and intersection points between the fitting plane and the data boundary; and fitting a polynomial surface so that all points in the point set will be close to the surface. In one aspect of the present invention, the rendering direction is set to be the same as the fitting plane, the surface is rendered, and the rendering result is projected to the fitting plane. The rendering mode can be selected from the group consisting of: multi-planar reformatting (MPR), maximum intensity projection (MIP), minimum intensity projection (MinIP), average intensity projection (AIP) and volume rendering (VR).

[0018] In one aspect of the invention, segmenting ribs and detecting and marking rib centerlines are performed via machine learning or deep learning techniques. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In the following figures:

[0020] Figure 1 is a diagram of an exemplary imaging system;

[0021] Figure 2 Views of the thoracic cavity according to a "fillet view" or "fishbone view" and a visceral cavity view are illustrated.

[0022] Figure 3 The rib centerline points in a single plane are illustrated;

[0023] Figure 4 illustrates visualization of a rib based on an oblique axis view corresponding to the fitted plane;

[0024] Figure 5 The figure shows the automatically detected part of the rib that is partially out of plane;

[0025] Figure 6 An enhanced view is illustrated;

[0026] Figure 7 illustrates an MPR view and an enhanced view according to some embodiments;

[0027] Figure 8 Surface fitting is illustrated;

[0028] Fig. 9 illustrates an enhanced view according to some embodiments;

[0029] Fig.10 is a flowchart representing some embodiments;

[0030] Fig.11 is a flow chart representing a method for enhancing rib plane visualization via a fusion scheme according to some embodiments; and

[0031] Fig.12 is a flow chart representing a method for enhancing rib plane visualization via a curved surface scheme according to some embodiments. DETAILED DESCRIPTION

[0032] Figure 1 An imaging system 100, such as a computed tomography (CT) imaging system, is illustrated. The imaging system 100 includes a generally fixed gantry 102 and a rotating gantry 104. The rotating gantry 104 is rotatably supported by the fixed gantry 102 and rotates relative to the fixed gantry 102 about a longitudinal axis or z-axis about an examination region 106.

[0033] A patient support 112, such as a couch, supports an object or examination subject, such as a human patient, in the examination region 106. The support 112 is configured to move the object or examination subject for loading, scanning, and unloading.

[0034] A radiation source 108, such as an X-ray tube, is rotatably supported by the rotating gantry 104. The radiation source 108 rotates with the rotating gantry 104 and emits radiation that passes through the examination region 106.

[0035] A radiation sensitive detector array 110 encloses an angular arc across the examination region 106 opposite the radiation source 108. The detector array 110 includes one or more rows of detectors extending along the z-axis direction, detects radiation traversing the examination region 106, and generates projection data indicative thereof.

[0036] A general purpose computing system or computer is used as an operator console 114 and includes (one or more) input devices 116 (such as a mouse, keyboard, etc.) and / or (one or more) output devices 120 (such as a display monitor, film camera, etc.). The console 114 allows the operator to control the operation of the system 100. This includes control of the imaging processing device 118. The imaging processing device 118 can receive data representing a three-dimensional (3D) diagnostic image generated by the imaging system 100, calculate a fitting plane corresponding to the rib centerline for each rib, respectively, generate a visualization of the rib, and generate an enhanced visualization when a rib portion is outside of a determined fitting plane. The user can switch between the original visualization of a given rib and the enhanced visualization of the rib.

[0037] It should be understood that the processing of the imaging processing device 118 can be implemented by (one or more) microprocessors that execute (one or more) computer-readable instructions encoded or embedded on computer-readable storage media (such as physical memory and other non-transient media). Additionally or alternatively, (one or more) microprocessors can execute (one or more) computer-readable instructions carried by carrier waves, signals and other transient media.

[0038] Figure 2 Various views of the thorax are illustrated according to a visceral cavity view 202 and a "fillet view" or "fishbone view" 204. As can be seen, these views allow a medical professional (such as a radiologist) to accurately examine the rib centerline in an expanded view (reformatted view) in which all ribs are straightened and placed in a unique position on the examination canvas. These views allow the entire thorax to be evaluated effectively, but they introduce new views that the radiologist needs to become familiar with and can show significant distortions and even hide pathologies such as fractures.

[0039] The presented novel visualization scheme is based on axial view plane adjustment. The axial plane is the view used by most radiologists to primarily interpret relevant imaging studies. Typically, most of the ribs lie within a single plane. To exploit this property, the user is shown a reformatting corresponding to the best view plane. Oblique axial views can be easily inspected by the radiologist. However, parts of the ribs may not lie within the plane and, therefore, may be overlooked.

[0040] Thus, the novel automatic enhancement scheme automatically detects out-of-plane portions to produce an enhanced view that allows inspection of the out-of-plane portions. Two different implementations of the enhanced view are provided, and the affected portions may be highlighted to allow the user to identify portions of the rib that were not part of the original visualization of the rib.

[0041] One enhanced view is provided by performing fusion of portions that are not located in the plane.Another enhancement is provided by generating a curved surface close to the oblique axial plane and the rib centerline point.

[0042] For each enhanced view, segmentation of the ribs is performed, the rib centerlines are extracted, and a fitting plane corresponding to the rib centerlines is calculated for each rib fitting plane.

[0043] Figure 3 The centerline points of the ribs in a single plane are illustrated. Depending on the segmentation, several different representations can be derived. Rib centerline visualization is a common representation derived from the segmentation. The segmentation and / or derived representations can be labeled according to anatomical labels (i.e., L1-12, R1-12). There are several mature methods for segmentation, labeling, and centerline extraction, and sufficient robustness has been successfully demonstrated. An example of such an approach is the application of a fully convolutional neural network (FCNN) to generate a probability map for detecting the first rib pair, the twelfth rib pair, and a set of all intermediate ribs. In the second stage, a centerline extraction algorithm is applied to this multi-label probability map. Finally, the different detections of the first rib and the twelfth rib allow the individual rib labels to be derived by classifying and counting the detected centerlines, respectively. The rib centerlines can be obtained using an interpolation scheme such as a spline.

[0044] Typically, most of the centerline points of the ribs will lie roughly within a single plane. The plane can be determined via a covariance analysis of the coordinates of the center points, where the eigenvector corresponding to the smallest eigenvalue of the covariance matrix corresponds to the plane normal vector. Since the ribs (especially at the vertebral junctions and close to the sternum) often show significant curvature, a weighted covariance matrix focused on the central portion of the ribs is beneficial to ensure that the calculated plane passes through most points.

[0045] The different weightings of the centerline points calculated for the plane fit are illustrated at 302. The lighter parts have lower weights to ensure focus on the main part of the rib. The plane fit through the rib is illustrated at 304, which is shown as a mesh.

[0046] For the fusion scheme, using the calculated plane fit, the out-of-plane parts can be automatically determined by calculating the distance d to the found plane for each center line. All points larger than a specified threshold T will be considered out-of-plane. Based on the determined distances, two main scenarios can be defined.

[0047] Figure 4 A first scenario is illustrated. In the first scenario, all points are located within the plane or close enough to the plane. By visualizing the oblique axial view corresponding to the determined plane fit, the ribs can be well evaluated. The visualization of the rib pairs 402 and 404 based on the oblique axial view corresponding to the fitted plane is shown in FIG. Figure 4 middle.

[0048] Figure 5 The second scenario is illustrated. The second scenario is a situation where some points are outside the plane. In this case, some parts of the rib cannot be evaluated using the tilt axis derived from the plane fit. In these cases, the user starts navigating to display the out-of-plane parts. However, the parts that were initially in-plane will disappear. This increases the complexity and reading time of the image slices. Therefore, in such scenarios where not all parts are in-plane, an "enhanced view" is implemented. The parts in the plane are displayed as before, but the parts outside the plane are projected into the plane. More precisely, since the ribs are bright structures, a maximum intensity projection can be performed locally. By smoothly transitioning this projection, sharp edges in the image are prevented.

[0049] The portion of the rib outside of the multi-planar reformatting (MPR) is automatically determined at 502. The larger change area along the rib is determined at 504, and the depth change of the MPR plane 506 is smoothly changed to ensure a smooth resulting image. The fusion is limited to the area of ​​the rib that is locally outside the plane. Therefore, most of the image is not affected. The fused portion can be highlighted, for example, via a color overlay or colored circle to identify to the user one or more areas that are not part of the original image.

[0050] Figure 6An enhanced view of an out of plane rib portion and an out of plane portion are illustrated. An MPR in which a rib portion is slightly out of plane and completely out of plane is shown at 602. A missed fracture is shown in the enhanced view at 604. The highlighting can be extended to the entire area that could potentially be considered to show the radius of influence. An indication of the change zone is shown at 606. This can be provided to the clinician or radiologist as a warning message, and the highlighting can be turned on and off by the clinician.

[0051] Another example of an MPR view and an enhanced view is Figure 7 The MPR view is shown at 702 and the enhanced view is shown at 704. The region at 706 of the enhanced view 704 illustrates the out-of-plane portion of the rib that is missing in the MPR view 702.

[0052] For the curved surface solution, it should be understood that there are an infinite number of surfaces that can intersect the centerline point, and given a series of surface equations, the closest surface may not be close to the fitting plane (the tilted axial plane). In order to ensure that the surface is close to this plane, it is necessary to expand the point set with points that lie on the plane. This limits any potential irregularities of the surface so that the image content will be more natural.

[0053] Polynomial equations are a way to construct and solve curved surfaces, and the curvature can be easily controlled by higher-order parts. The accuracy of the surface can be improved by adding higher-order parts, and the curvature can be restricted by adding penalties to the higher-order parts. The order of the polynomial equation can be determined by limiting the average distance between the set of points and the calculated surface.

[0054] Figure 8 Surface fitting is illustrated. A fitted plane where portions of points are locally outside the plane is shown at 802. An enhanced view via surface fitting is shown at 804. The fitted polynomial surface is close to the original fitted plane, and the distance between the points and the surface decreases.

[0055] Fig. 9 An enhanced view with surface fitting is illustrated. At 902, an MPR is shown where a portion of a rib is locally out of plane. An enhanced view is shown at 904. The enhanced view shows the out-of-plane portion in region 906 that is missing in the MPR 902.

[0056] Several different rendering modes can be applied, such as Maximum Intensity Projection (MIP), Minimum Intensity Projection (MinIP), Average Intensity Projection (AIP), and Volume Rendering (VR), verifying that the entire image shares the same thickness.

[0057] For both the fusion method and the curved surface method, a plane fit is calculated for each rib separately in order to efficiently navigate through the entire thoracic cavity. The left and right ribs can be displayed side by side, as shown in Figure 1. Figure 4 As illustrated. Overview images can be shown for all 12 rib pairs in one stacked overview. Interaction can be enabled to click from one rib pair to the next. For each rib, individual navigation is possible (preferably parallel to the axis and limited to the ribs).

[0058] Fig.10 is a flow chart representing a method of enhancing rib plane visualization according to some embodiments.At 1002, data representing a three-dimensional (3D) diagnostic image including ribs of a subject (eg, a human patient) is received.

[0059] At 1004, the ribs are segmented based on the received data, the data representing a 3D diagnostic image. Segmenting the ribs can be performed manually, semi-automatically, or by applying an automated machine learning method such as a fully convolutional neural network (FCNN) to generate a probability map for detecting the first rib pair, the twelfth rib pair, and the set of all intermediate ribs. Next, a centerline extraction algorithm is applied to the multi-label probability map. The rib centerlines can be obtained using cubic spline interpolation. Finally, the different detections of the first rib and the twelfth rib, respectively, allow the individual rib labels to be derived by simple classification and counting of the detected centerlines.

[0060] At 1006, rib centerlines are detected and labeled from the rib segmentation. The segmentation of the ribs may be performed via machine learning or deep learning techniques.

[0061] At 1008, a fitted plane corresponding to the rib centerline is calculated for each rib, respectively. An out-of-plane rib portion is determined for each rib at 1010. At 1012, a visualization of the ribs is generated. The visualization at 1012 may be generated for each rib individually or as a rib pair. An enhanced visualization for the out-of-plane rib portion is generated at 1014.

[0062] Fig.11 is a flow chart representing a method for enhancing rib plane visualization via a fusion scheme according to some embodiments. At 1102, the distance to the fitted plane is calculated for each rib centerline point. That is, for each centerline, the distance d to the found plane is calculated. In the case where all points are within or close to the plane, the rib(s) can be evaluated to visualize an oblique axial view corresponding to the determined plane. This is done in Figure 4 It is shown as 402.

[0063] For cases where some points are outside the plane, the associated out-of-plane portion of the rib cannot be evaluated using the tilt axis derived from the fitted plane. Using the calculated fitted plane, the out-of-plane rib portion can be automatically determined. The automatically determined out-of-plane rib portion is projected onto the visualization of the rib at 1104 and is viewed as fused with the in-plane portion of the rib, as shown in FIG. Figure 4 402. The fusion is limited to the region where the ribs are locally out of plane, leaving most of the image area including the region where the ribs are in plane unaffected.

[0064] At 1106, out-of-plane rib portions can be highlighted in the enhanced view to identify rib portions that are not part of the original visualization. At 1108, the radius of influence can be highlighted in the enhanced visualization of the ribs. At 1110, the user can toggle between the original visualization of the ribs and the enhanced visualization of the ribs.

[0065] Fig.12 is a flow chart representing a method for enhancing rib plane visualization via a curved surface scheme according to some embodiments. At 1202, a point set is established including all centerline points of the rib and intersection points between the fitted plane and the data boundary.

[0066] Given a series of surface equations, the closest surface may not be close to the fitted plane, given that there are countless surfaces that can cross the centerline point. To ensure that the surface is close to the plane, the point set can be extended with points that lie on the plane. Polynomial equations are an effective way to construct curved surfaces. At 1204, the polynomial surface is fitted so that all points in the point set will be close to the surface.

[0067] The order of the polynomial equation is determined by limiting the average distance between the point set and the calculated surface. The accuracy of the surface can be improved by adding higher order parts. At 1206, the drawing direction is set to be the same as the fitted planar surface. Fig. 9 MPR rendering and enhanced views according to the curved surface scheme are shown. Different rendering modes including MIP, MinIP, AIP and VR can also be applied. At 1208, the rendering results are projected to the fitting plane.

[0068] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments.

[0069] Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

[0070] In the claims, the word "comprising" does not exclude other elements or steps, and the word "a" or "an" does not exclude a plurality.

[0071] A single processor, device or other unit may fulfill the functions of several items recited in the claims.The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0072] Operations such as collecting, determining, obtaining, outputting, providing, storing or storing, calculating, simulating, receiving, warning and stopping may be implemented as program code elements of a computer program and / or dedicated hardware.

[0073] The computer program may be stored / distributed on suitable media, such as optical storage media or solid-state media provided together with or as part of other hardware, but the computer program may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems.

Claims

1. A method for enhancing rib plane visualization, the method comprising: receiving data representing a three-dimensional diagnostic image including one or more ribs of a subject; segmenting the one or more ribs based on the received data representing the three-dimensional diagnostic image; Detect and mark rib centerlines based on rib segmentation; Calculating a fitting plane corresponding to the center line of the rib for each rib respectively; Determine the out-of-plane rib portion for each rib; generating a visualization of the ribs individually or as rib pairs; and Generates enhanced visualization of out-of-plane rib sections.

2. The method according to claim 1, wherein: Generating an enhanced visualization for out-of-plane rib portions includes calculating a distance to a fitting plane for each rib centerline point, and wherein rib portions outside the fitting plane are projected onto the visualization of the ribs, fused with in-plane rib portions.

3. The method according to claim 2, wherein: The out-of-plane rib portions are highlighted to identify rib portions that were not part of the original visualization of the rib. 4 . The method of claim 2 , further comprising switching between an original visualization of the rib and the enhanced visualization of the rib. 5 . The method of claim 2 , further comprising highlighting a radius of influence in the enhanced visualization of the rib.

6. The method according to claim 1, wherein: Generating enhanced visualization for the out-of-plane rib portion includes: establishing a point set including all centerline points of the rib and intersection points between the fitted plane and the data boundary; and fitting a polynomial surface so that all points in the point set will be close to the surface.

7. The method according to claim 6, further comprising: The drawing direction is set to be the same as the fitting plane, the surface is drawn, and the drawing result is projected onto the fitting plane.

8. The method according to claim 7, wherein: The rendering mode is selected from the group consisting of: multiplanar reformatting (MPR), maximum intensity projection (MIP), minimum intensity projection (MinIP), average intensity projection (AIP), and volume rendering (VR).

9. The method according to claim 1, wherein: Segmenting the ribs and detecting and marking rib centerlines are performed via machine learning or deep learning techniques.

10. An imaging device comprising: a memory configured to store computer-executable instructions; as well as at least one processor configured to execute the computer executable instructions to cause the imaging device to: receiving data representing a three-dimensional diagnostic image including one or more ribs of a subject; segmenting the one or more ribs based on the received data representing the three-dimensional diagnostic image; Detect and mark rib centerlines based on rib segmentation; Calculating a fitting plane corresponding to the center line of the rib for each rib respectively; Determine the out-of-plane rib portion for each rib; generating a visualization of the ribs individually or as rib pairs; and Generates enhanced visualization of out-of-plane rib sections.

11. The imaging device according to claim 10, wherein: The processor is configured to: generate an enhanced visualization for out-of-plane rib portions by calculating the distance to the fitting plane for each rib centerline point; and project the rib portions outside the fitting plane onto the visualization of the ribs, appearing fused with the in-plane rib portions.

12. The imaging device according to claim 11, wherein: The processor is configured to highlight the out-of-plane rib portions to identify rib portions that were not part of an original visualization of the rib.

13. The imaging device according to claim 11, wherein: The processor is further configured to switch between the original visualization of the rib or the pair of ribs and the enhanced visualization of the rib or the pair of ribs based on user input.

14. The imaging device according to claim 11, wherein: The processor is further configured to highlight a radius of influence in the enhanced visualization of the rib or the pair of ribs based on user input.

15. The imaging device according to claim 10, wherein: The processor is configured to generate an enhanced visualization for the out-of-plane rib portion by: establishing a point set including all centerline points of the rib and intersection points between the fitted plane and the data boundary; and fitting a polynomial surface so that all points in the point set will be close to the surface.

16. The imaging device according to claim 15, wherein: The processor is further configured to: set a drawing direction to be the same as the fitting plane, draw the surface, and project the drawing result onto the fitting plane.

17. The imaging device according to claim 16, wherein: The processor is configured to select a rendering mode from the group consisting of: multi-planar reformatting (MPR), maximum intensity projection (MIP), minimum intensity projection (MinIP), average intensity projection (AIP), and volume rendering (VR).

18. The imaging device according to claim 10, wherein: Segmenting the ribs and detecting and marking the rib centerlines are performed via machine learning or deep learning techniques.

19. A non-transitory computer readable medium having stored thereon instructions for causing a processing circuit to perform a process comprising: receiving data representing a three-dimensional diagnostic image including one or more ribs of a subject; segmenting the one or more ribs based on the received three-dimensional diagnostic image data; Detect and mark rib centerlines based on rib segmentation; Calculating a fitting plane corresponding to the center line of the rib for each rib respectively; Determine the out-of-plane rib portion for each rib; generating a visualization of the ribs individually or as rib pairs; and Generates enhanced visualization of out-of-plane rib sections.

Images